Due to its asymptomatic nature and lack of methods for early detection, > 80% of pancreatic cancer (PC) patients present with an unresectable primary tumor with distant metastasis at the time of diagnosis. While the overall 5 year survival rate of pancreatic cancer is dismal, significantly better outcomes have been reported for smaller tumors detected at an earlier stage. Slow development of PC in conjunction with the better curative response of patients with early disease underscore the need of early detection of pancreatic cancer. While there is no consensus on the diagnostic approaches for early detection of PC, in light of its rare prevalence, there is a growing agreement to target high-risk individual for early diagnosis. Individuals with familial PC and patients harboring cystic lesions in the pancreas are considered be the two well defined high-risk groups likely to develop pancreatic cancer. Pancreatic cystic lesions, earlier considered to be rare, are increasingly being recognized due to increased number of individuals being subjected to diagnostic imaging; however, their exact prevalence is unknown. These cystic pancreatic lesions have variable malignant potential: while mucinous cystic neoplasms (MCNs) and intraductal pancreatic mucinous neoplasms (IPMNs) have a high probability of developing into malignant lesions, serous cystic neoplasms (SCNs) are considered benign. Despite the critical need, accurate discrimination between high- and low-risk cystic lesions is challenging due to their symptomatic and radiographic similarities. Although cytologic examination of endoscopic ultrasound (EUS) guided fine needle aspirates (FNAs) has emerged as an indispensable part of presurgical evaluation, in practice, 50%-60% of such analyses are inconclusive and unreliable in discriminating between serous and mucinous lesions. Using anti-MUC4 monoclonal antibody 8G7 generated by our group several studies have established that cell surface mucin MUC4 is promising prognostic and diagnostic biomarker. MUC4 expression increases progressively in precursor PanIN lesions and is detectable in EUS FNAs. The central hypothesis of this proposal is that the detection of MUC4 in pancreatic tissues is positively correlated with the presence of already existing pancreatic cancer or precancerous lesions and thus could be a powerful tool for the early diagnosis and for predicting the prognosis of patients affected with this lethal disease. Two specific aims are proposed. In Aim 1 (Phase I) using archived FNAs and matched resected specimens, we will demonstrate our ability to successfully determine MUC4 expression in cystic pancreatic lesions and establish that MUC4 expression can predict the occurrence of occult malignancy or preneoplasatic lesions that are otherwise undetectable by conventional methods. Future studies will validate the significance of MUC4 immunostaining in a multi-center trial and determine how MUC4 expression correlates with the clinical outcome of the solid/cystic pancreatic diseases. Overall, the proposal will serve as a platform to determine if there is a potential role of MUC4 in clinical decision making in the context of pancreatic cystic lesions and PC.